I am following https://stackoverflow.com/a/77851176/243031 to create my model optional.
I created function to get base class annotations.
def get_annotations(main_cls):
ret_val = main_cls.__annotations__
for base_cls in main_cls.__bases__:
if base_cls != BaseModel:
ret_val.update(get_annotations(base_cls))
return ret_val
and created optional model as
OptionalClientModel = create_model(
"OptionalClientModel",
**{k: (Optional[v], None) for k, v in get_annotations(ClientModel).items()})
The original classes are as below
from typing import Annotated
from bson import ObjectId
from pydantic import Field
from pydantic import EmailStr
from pydantic import BaseModel
from pydantic import BeforeValidator
from pydantic import ConfigDict
from pydantic import AwareDatetime
from pydantic import field_validator
# Represents an ObjectId field in the database.
# It will be represented as a `str` on the model so that it can
# be serialized to JSON.
PyObjectId = Annotated[str, BeforeValidator(str)]
class DBTableBase(BaseModel):
# The primary key for the Table, stored as a `str` on the instance.
# This will be aliased to `_id` when sent to MongoDB,
# but provided as `id` in the API requests and responses.
id: PyObjectId | None = Field(alias="_id",
serialization_alias="id",
default=None)
model_config = ConfigDict(
json_encoders={ObjectId: str},
json_schema_extra={
"example": {
"id": "BSON_ID"
}
},
)
class ClientModel(DBTableBase):
first_name: str
last_name: str
When I want model with all optional value, I can use OptionalClientModel
.
The issue is, id
in OptionalClientModel
has no alias.
How to create optional with alias?
With the following approach you can pass the FieldInfo
objects from the original model to the new dynamically created model while giving the annotation an extra | None
.
OptionalClientModel = create_model(
"OptionalClientModel",
**{k: (v.annotation | None, v) for k, v in ClientModel.model_fields.items()})
Edit:
To set a default value of None
as well you have to mutate the FieldInfo
object before passing them to the create_model
function. You also may have to copy the FieldInfo
object on beforehand. Normally it wouldn't change anything but I am almost sure (didn't test it) that if you are calling model_rebuild
on the original model somewhere the changes on the FieldInfo
objects will be reflected on the rebuilt pydantic core schema.
field_infos = {}
for field_name, field_info in ClientModel.model_fields.items():
field_info_copy = copy(field_info)
field_info_copy.default = None
field_infos[field_name] = field_info_copy
OptionalClientModel = create_model(
"OptionalClientModel", **{k: (v.annotation | None, v) for k, v in field_infos.items()}
)